Formation control of wheeled robots with vision-based position measurement

Many applications require multiple mobile robots to move with a common velocity and at fixed relative distances. We present a time-invariant, state-feedback control law and a novel vision-based pose reconstruction system that allows one differential drive robot to follow another at a constant relative distance. The control law does not require measurement or estimation of the leader robot velocity and has tunable parameters that allows one to prioritize the error bounds of the desired states. The proposed pose reconstruction algorithm is computationally inexpensive and reliable. We present experimental results on two iRobot Create robots showing the performance of the controller and vision algorithm.